Abstract
A computational model of resilient T cells predicts immunotherapy response via transcriptomic data.
Major Finding: A computational model of resilient T cells predicts immunotherapy response via transcriptomic data.
Concept: The model identifies genes such as FIBP that affect T-cell persistence despite immunosuppressive signals.
Impact: This work suggests novel therapeutic targets that may functionally dictate patient response.
Despite the clinical success of immunotherapy for a subset of patients with different cancer types, durable response to immunotherapy often remains limited by an immunosuppressive tumor microenvironment that prevents optimal T-cell effector function. To explore key regulators that dictate T-cell response to immunosuppressive signals, Zhang, Vu, and colleagues integrated single-cell RNA-sequencing (scRNA-seq) data from 168 tumors representing 19 cancer types to develop Tres, a computational model that identifies transcriptomic signatures of tumor-resilient T cells, defined as T cells that continue to proliferate under immunosuppressive pressures. Focusing on TGFβ1, TRAIL, and prostaglandin E2, the signaling activities of these immunosuppressive cytokines (predicted by CytoSig) were found to negatively correlate with T-cell proliferation levels, and the Tres model identified genes that significantly mitigated this negative correlation in the majority of tumor samples. Trained on scRNA-seq data from immunotherapy-naïve tumor samples, Tres successfully predicted clinical response to immune checkpoint blockade, adoptive T-cell transfer, and other immunotherapies based on bulk RNA-seq data from pretreatment tumors, infusion products, and premanufacture samples for cell therapies. Tres identified acidic fibroblast growth factor intracellular-binding protein (FIBP) as one of the top marker genes whose high expression correlated with poor T-cell resilience and predicted clinical response to immunotherapy, while genetic studies supported a functional role for FIBP, as FIBP knockout in human T cells enhanced cytokine release and killing of cognate tumor cells in vitro. In addition, adoptive transfer of Fibp knockout T cells into tumor-bearing mice significantly reduced tumor growth as compared to mice treated with control T cells. Transcriptomic analysis of FIBP suppression revealed the role of FIBP in negatively regulating T-cell efficacy through metabolic effects, with FIBP knockdown leading to downregulation of cholesterol metabolism via decreased expression of multiple cholesterol synthesis enzymes, cholesterol uptake receptors, and the master transcription factor SREBF2. In summary, this study describes a computational model that predicts immunotherapy response, highlighting FIBP as a promising target to enhance T-cell efficacy in the presence of immunosuppressive signaling.
Note:Research Watch is written by Cancer Discovery editorial staff. Readers are encouraged to consult the original articles for full details. For more Research Watch, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/CDNews.
RSS Feeds